colleges are more like cruise ships and bars than kindergartens and elementary schools

The following post is co-authored with Jessica Calarco.

In the wake of the coronavirus pandemic, policymakers at the national, state, and local level are scrambling to decide what can reopen while limiting the virus’s spread. In some sense, we can think of the overall rate of infection of the virus as a kind of budget constraint. As long as the rate of spread is kept below 1, the virus is under control. If too much opens up, and the rate goes above 1, the virus will begin to overwhelm the healthcare system, as we’re currently seeing in Arizona, South Carolina, and Texas. Keeping everything closed would certainly help avoid that outcome. But keeping everything closed also comes with costs to the economic and social/emotional well-being of individuals, organizations, and society as a whole. Thus, the question that policymakers face – at least assuming they want to avoid massive, unnecessary deaths – is what to reopen given that some things have to stay closed? Or, put differently, which institutions are so critical to society that reopening them for in-person use is ultimately worth the risk? 

To answer that question, let’s consider a spectrum of organizations, ranked by the benefits of reopening versus the associated increase in risk of spread. At one extreme, we have bars and cruise ships. Sure, reopening them has benefits for patrons, but their benefits for society are more limited, and reopening them comes with huge risks that are difficult to contain. How do you drink a beer while wearing a mask? What’s the point of being on a cruise ship if you can’t eat in the restaurants or even leave your room? On another extreme, we have nursing homes and hospitals. There are risks to keeping these institutions open, but those risks can be managed with careful protocols and plenty of PPE (personal protective equipment). In terms of benefits, these institutions are so critical that it would be almost inconceivable for them to close. Could we really tell people to find somewhere else to live or fix their own broken legs?

This approach is useful for considering whether to reopen schools. Essentially, we should be asking whether schools are more like bars and cruise ships, or more like hospitals and nursing homes. And we should be willing to consider the possibility that different school types—preschools, elementary schools, middle schools, high schools, and higher ed—might fall at different ends.

Ultimately, we would argue that preschools and elementary schools fall closer to nursing homes and hospitals on our risk vs. benefit spectrum, while high schools and colleges are closer to bars and cruise ships, and middle schools are somewhere in between. These distinctions, we should note, are not based on the educational benefits these different institutions provide for individual students. There is clear evidence — from preschool to grad school — that individuals benefit from more education. Nor are these distinctions based solely on the risks that attending school would pose to individual students (certainly, young children can get the virus, but there’s also evidence that the risks of catching and spreading it increase with age). Instead, these distinctions are based on the extent to which the benefits of these institutions (as institutions of learning and as institutions that serve other social functions, as well) are contingent on in-person instruction and how easily the risks of reopening them can be managed to reduce the spread of harm.

Preschools and elementary schools play a critical role in educating our nation’s young children, and there is reason to suspect that those academic benefits are heavily dependent on in-person instruction. A new paper by economist Carycruz Bueno finds that, for K-12 students, attending virtual schools leads to significant reductions in achievement test scores and even a 10 percent reduction in students’ chances of graduating from high school. Those negative outcomes are even more pronounced among students in younger grades. Bueno’s study focused on schools designed to be virtual, not in-person schools scrambling to adapt to the pandemic by switching to online instruction. So there is reason to suspect that learning losses will be even more pronounced if preschool and elementary schools remain closed. Further evidence for that possibility can be found in research showing that young children need a tremendous amount of hands-on support with learning, especially if they cannot yet read fluently.  Most families do not have a parent who can provide full-time, hands-on support. Thus, if schools for young children do not reopen, the learning losses will be large, and will disproportionately affect students from families with the fewest resources.

In addition to their role as institutions of learning, preschools and elementary schools also serve other critical social functions, and those non-academic benefits are contingent on in-person instruction, as well. For working families and especially working mothers, preschools and elementary schools operate as childcare providers, allowing parents and especially mothers to work. For parents who have to work outside the home during the pandemic, keeping preschools and elementary schools closed may mean deciding between quitting their jobs, putting grandparents’ health at risk by asking them to provide full time care, or risking having their children taken away (because many states make it illegal for young children to stay home alone). Even for parents who can work from home, evidence suggests that the challenges of supporting children’s learning while working full time are just too great for many families to bear. Recent research by sociologists Caitlyn Collins and colleagues, example, has shown that the pandemic significantly reduced work hours among parents of young children and that those burdens disproportionately impact mothers. Not reopening preschools and elementary schools, or only reopening them part-time, runs the risk of exacerbating long-standing workplace inequalities between men and women and between women with and without kids.  

In terms of managing the risks of reopening, the evidence actually favors in-person preschools and elementary schools, as well. Certainly, it is not easy to get young children to wear masks and keep their distance. But the structure of preschools and elementary schools may actually help in terms of limiting risks. In these schools, kids are typically with the same group of students and the same teacher all day. Students aren’t typically switching classrooms, and there aren’t usually extracurricular activities after school. If, following CDC recommendations, schools have students eat lunch in classrooms, stagger recesses, and add extra buses, they can keep contact between students from different classrooms fairly low. If high schools close, as some districts have proposed, that extra space can also be used to make K-5 classes smaller and spread students out even more. Of course, there may be some preschool and elementary students, especially those in high risk groups or with family members in high risk groups, for whom in-person instruction is too dangerous. For those students, however, and as some districts have proposed, it is possible to offer online instruction as an opt-in alternative, rather than forcing families to figure out an alternate solution on their own. 

These changes will be costly. Preschools and elementary schools will need to hire more teachers to keep class sizes small, and they should arguably increase teacher pay to account for the added risk. Schools will also need more buses and bus drivers, more staff and supplies for regular cleanings, upgraded ventilation systems and plumbing to ensure that there’s clean air to breathe and hot water for washing hands, and plenty of soap, hand sanitizer, and masks to go around. Given that local and state budgets are already strapped, putting these protections in place will require a massive investment of federal dollars. But that investment is necessary not only for reaping the benefits outlined above but also for reducing the potential for the virus to spread among teachers and staff, who are (because of their age) at greater risk.

These same considerations operate in reverse for high schools. Our point is not that high school online would be as academically or socially beneficial for most students as high school in-person; the evidence suggests that it will be worse. But the difference between high school online and elementary school online is large. High schoolers are at least capable of learning online, especially if schools are able to provide them with the appropriate technology, including reliable internet access. Parents who have to work can leave their high school-aged children at home, as high schools are not providing childcare in the same fashion as preschools and elementary schools. The structure of high school also makes it more difficult to limit the spread of the virus, as students are typically moving between many different classrooms with different specialist teachers, as well as a host of extracurricular activities from athletics to band. These activities will be risky and would likely have to be curtailed even if in-person instruction resumes. Of course, there may be some high school students who, because of special needs or unstable home environments, may be unable to learn online. For those students, schools could provide the option of in-person instruction, with the online default making that option safer for those students and teachers involved. 

For high schools, going virtual might ultimately be less costly than opening (safely) in person, but it will still come with costs, and those costs will also require a substantial public investment of funds. Schools will need to ensure that all students have access to their own reliable laptops or tablets, as well as a reliable high speed internet connection at home. In rural areas, this will also require adding more infrastructure, like laying internet cables and extending those connections to individual homes. Schools will also need to support teachers in developing effective plans for online instruction, ideally with reduced class sizes, more support staff (to assist students who need extra attention), and higher teacher pay. If schools give some students the option of in-person attendance, they will also need additional, well-compensated teachers and support staff for those students, as well. 

For residential colleges and universities, the situation falls even further on the spectrum away from the preschools and elementary schools. Certainly, in-person instruction is likely to be more effective for students’ learning than instruction delivered online. That said, classroom time forms a relatively small part of a college student’s week, perhaps 12-18 in-person hours. The rest of their academic time is already spent working independently or in small groups – doing course readings and problem sets, writing papers, and so on. With the exception of a small number of hands-on classes (e.g., labs, art and performance classes, and for-credit athletic activities), most of the work associated with most college classes is already done outside of the classroom. 

Like preschools and elementary schools, colleges and universities also serve non-academic functions, but the benefits of those functions are less tied to in-person instruction. Colleges and universities, for example, serve as hubs for research and innovation. While some of that work requires students’ on-campus presence (e.g., in labs), much of it is done by individual faculty members and graduate students, with limited undergraduate involvement. This kind of research activity, in turn, may be even easier to complete safely if faculty members and graduate students are not regularly coming into contact with large numbers of undergraduate students. 

To that end, the structure of residential colleges and universities will make it very difficult for them to open safely if large numbers of students are involved. Even more so than high schoolers, college students take a diverse mix of courses which in turn leads to a “small-world” network structure, as research by sociologists Kim Weeden and Benjamin Cornwell shows. Even eliminating large in-person lectures is not enough to disconnect the network, meaning that students are potentially at risk of infection from almost everyone else on campus just through their co-presence in the classroom. That degree of network overlap between students is further amplified by the centrality of non-academic activities on college campuses. Outside of the classroom, college students (who are almost all adults) engage in a wide-range of activities, from working jobs to taking care of family, as well as engaging in academic, athletic, and social extracurriculars.  Of course these risks can be mitigated – moving large lectures online, putting small classes in large rooms, and canceling many extracurricular activities. But doing so may further undermine the already relatively small advantage of in-person teaching over online courses (again, as compared to the differential for elementary schoolers). Furthermore, and given that many students opt to attend residential colleges because of the cruise ship-like amenities, canceling extracurricular activities and closing dining halls and stadiums and libraries and bars would ultimately negate the social benefits students hope to gain. Essentially, a safely reopened residential college experience loses much of its allure and value. 

Now, colleges and universities may need to make some accommodations for students who are unable to learn effectively online. Students without stable home environments will need a place to stay and ways to get food and other services delivered to them. Students without reliable laptops and internet access will need help accessing those technologies and may also need to live on campus. And students who are highly involved in research requiring on-campus resources may also need to be in residence, as well. By making online learning the default, however, it will be substantially easier to focus on maximizing the educational benefits to students while minimizing the risks to students, faculty, and staff. 

Of course, and as with moving high schools online, moving college classes online will also require a substantial investment of federal funds. Unlike public K-12 schools, which are funded almost exclusively by local, state, and federal tax dollars, most colleges and universities are funded primarily by students and their families. Universities rely on students for money from tuition and also from all the “extras” that students pay for when they live on campus, like room and board, meal plans, parking, and more. If residential colleges and universities close, students won’t be paying for all those “extras,” and some students may opt to take a gap year or complete their schooling at less expensive community colleges or fully online schools. To prevent massive financial losses, and to prevent the wave of (forever) school closures it might cause, colleges and universities will need a huge investment of federal funds. That includes money to offset the losses from tuition and the losses from “extras” that students aren’t buying if they’re not on campus. And it also includes resources to support instructors in reworking courses to be as effective as possible online.

Some universities have been very upfront about these challenges in their fall plans. For example, UMass-Amherst’s plan calls for almost entirely online classes, with dorms open for those students who feel they would benefit from being on-campus (perhaps because their home lacks reliable internet access, or a place to study). The plan’s key points end with the warning: “Life on campus will not be anything resembling normal college life.” And the Cal State system, which has a mix of residential students and commuters, declared in mid-May that they would be principally online in the fall. But most – 60% of the more than 1,000 tracked by the Chronicle for Higher Education – currently plan an in-person fall. 

Some universities have also been open about the primacy of financial considerations in driving their fall 2020 plans. Essentially, these schools have acknowledged that, when it comes to funding, universities are more like bars or cruise ships than public K-12 schools: their financial health depends on reopening. Notably, when Brown University President Paxson argued in her widely-circulated NYT op-ed in April that college campuses “must” re-open, she cited primarily financial pressures: 

Most colleges and universities are tuition dependent. Remaining closed in the fall means losing as much as half of our revenue.

This loss, only a part of which might be recouped through online courses, would be catastrophic, especially for the many institutions that were in precarious financial positions before the pandemic. It’s not a question of whether institutions will be forced to permanently close, it’s how many.

For this reason – along with the spillover economic benefits to college towns and businesses – Paxson argued that reopening college campuses should be “a national priority.” 

And perhaps it should. But how high a priority? More than cruise ships, perhaps. But more than K-12? More than preschool? We think not. Opening college campuses for on-campus instruction involves more risks for fewer benefits. To counter the risk of colleges closing, the federal government could provide funds to keep colleges afloat (directly, or indirectly by providing desperately needed to states who can in turn fund public higher-ed), the same as it has done for (some) businesses which have closed or reduced operations for public health reasons, and which other countries have done much more extensively and successfully. The closure of hundreds of colleges would be a tragedy, but it’s one that’s preventable without reopening college campuses. 

In contrast, the main costs of keeping preschools and elementary schools closed are to students and their parents, not to the schools themselves. These schools will need more resources – including especially to support students whose health prevents them from attending in-person and to make the teaching environment as safe as possible for teachers and staff – but these resources will directly support their work of providing essential educational benefits and childcare. 

Evidence from other countries suggests that preschools and elementary schools can be reopened safely. In many countries where the virus is overall under control, reopening these types of schools has not led to major increases in infections. Doing so required making some changes to how schools functioned in line with those discussed above, and responding quickly to outbreaks (enabled by adequate testing and resources for contact-tracing and supporting those who get sick), but these changes did not dramatically undermine the capacity of schools to provide education and childcare. 

Of course, whether states and local districts can follow this lead will depend, in part, on where and how the virus continues to spread. In some states, like Arizona, South Carolina, and Texas, the infection rates are so high and rising so rapidly that it may ultimately be unsafe to open anything, at least in the short term. Instead, and in the interest of staying within their infection “budget,” those states will need to shut down everything they possibly can as fast as they can, much as New York was forced to do in March. Other states like Michigan and Rhode Island, however, are hovering near that budget threshold, with just enough closed, and just enough mask usage and physical distancing, to keep the spread of the virus in check, but without a lot of wiggle room. 

In these states, we can reasonably ask, which institutions should we prioritize for reopening, and which others will have to remain closed? It’s easy to make the case that hospitals and nursing homes should stay open. And it’s easy to make the case that bars and cruise ships stay closed–the economic and social benefits are simply not worth the level of risk (as many have noted). But what about all the institutions that fall between bars and nursing homes on our risk/benefit spectrum? If we can’t safely open all levels of education, what should come first? Faced with that choice, we should prioritize reopening preschools and elementary schools while keeping high schools and college campuses closed. And the federal government should step in to ensure that the former effort is funded well enough to be done with minimal risks to students, teachers, and staff and that the latter can be done without threatening the long-term viability of public higher ed.